INDEX

Explanations

phrases related to geopolitical entities, specifically the United States

oai_token-act-pair · gpt-3.5-turbo

references to the United States

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

[|

-0.75

NULL

-0.69

%%

-0.68

ttp

-0.67

inventoryQuantity

-0.67

ï¸ı

-0.67

BOOK

-0.66

 Wink

-0.66

 inacc

-0.65

 refres

-0.64

POSITIVE LOGITS

 States

1.37

 Nations

1.20

 Kingdom

1.09

 STATES

1.03

States

0.99

 Federation

0.91

 Methodist

0.91

 Airlines

0.78

 Republic

0.77

 Continental

0.76

Activations Density 0.030%

phrases related to geopolitical entities, specifically the United States

references to the United States

No Comments

No Known Activations